U.S. patent application number 11/094250 was filed with the patent office on 2005-08-04 for interactive dialog-based training method.
This patent application is currently assigned to SBC Properties, L.P.. Invention is credited to Glaspie, Marie Jeanine, Koehler, Robert Kevin, Schatzlein, Gregory Scott.
Application Number | 20050170326 11/094250 |
Document ID | / |
Family ID | 27732830 |
Filed Date | 2005-08-04 |
United States Patent
Application |
20050170326 |
Kind Code |
A1 |
Koehler, Robert Kevin ; et
al. |
August 4, 2005 |
Interactive dialog-based training method
Abstract
A computer readable medium stores a computer program for
interactive training based on a simulated customer interaction. The
computer readable medium includes a dialog code segment that
provides a predetermined customer portion of a dialog segment from
the simulated customer interaction to a trainee. The dialog segment
includes the customer portion and at least one predetermined
keyword associated with the customer portion. The computer readable
medium further includes a response code segment, which receives a
response to the customer portion from the trainee, and a
determining code segment, which determines whether the response
includes the at least one keyword.
Inventors: |
Koehler, Robert Kevin;
(Chicago, IL) ; Schatzlein, Gregory Scott; (St.
Charles, IL) ; Glaspie, Marie Jeanine; (Indianapolis,
IN) |
Correspondence
Address: |
GREENBLUM & BERNSTEIN, P.L.C.
1950 ROLAND CLARKE PLACE
RESTON
VA
20191
US
|
Assignee: |
SBC Properties, L.P.
Reno
NV
|
Family ID: |
27732830 |
Appl. No.: |
11/094250 |
Filed: |
March 31, 2005 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11094250 |
Mar 31, 2005 |
|
|
|
10078429 |
Feb 21, 2002 |
|
|
|
Current U.S.
Class: |
434/350 ;
434/323; 434/362 |
Current CPC
Class: |
G09B 5/00 20130101; G09B
19/00 20130101; H04M 3/51 20130101 |
Class at
Publication: |
434/350 ;
434/362; 434/323 |
International
Class: |
F41G 003/26 |
Claims
What is claimed:
1. An interactive training method based on a simulated customer
interaction, the method comprising: providing a predetermined
customer portion of a dialog segment from the simulated customer
interaction to a trainee via a graphical user interface, the dialog
segment comprising the customer portion and at least one
predetermined keyword associated with the customer portion;
receiving a response to the customer portion from the trainee via
the graphical user interface; and determining whether the response
includes the at least one keyword.
2. The interactive training method according to claim 1, further
comprising: providing a second predetermined customer portion of
the dialog segment to the trainee via the graphical user interface,
the dialog segment further comprising a second predetermined
keyword associated with the second customer portion; receiving a
second response to the second customer portion from the trainee via
the graphical user interface; and determining whether the second
response includes the second keyword.
3. The interactive training method according to claim 1, in which
providing the predetermined customer portion comprises playing at
least a portion of an audio file at the graphical user
interface.
4. The interactive training method according to claim 3, in which
receiving the response to the customer portion comprises receiving
and recording a verbal response from the trainee.
5. The interactive training method according to claim 1, further
comprising: recording the response as an audio file; and playing
the audio file in response to a request received from the graphical
user interface.
6. The interactive training method according to claim 1, further
comprising: providing instruction to the trainee via the graphical
user interface based on at least the response to the customer
portion.
7. The interactive training method according to claim 6, further
comprising: determining a level of instruction to provide to the
trainee prior to providing the instruction, the level of
instruction being based on at least a determination of performance
by the trainee.
8. A computer readable medium for storing a computer program for
interactive training based on a simulated customer interaction, the
computer readable medium comprising: a dialog code segment that
provides a predetermined customer portion of a dialog segment from
the simulated customer interaction to a trainee, the dialog segment
comprising the customer portion and at least one predetermined
keyword associated with the customer portion; a response code
segment that receives a response to the customer portion from the
trainee; and a determining code segment that determines whether the
response includes the at least one keyword.
9. The computer readable medium according to claim 8, in which: the
dialog code segment provides a second predetermined customer
portion of the dialog segment to the trainee, the dialog segment
further comprising a second predetermined keyword associated with
the second customer portion; the response code segment receives a
second response to the second customer portion from the trainee;
and the determining code segment determines whether the second
response includes the second keyword.
10. The computer readable medium according to claim 8, further
comprising: a display code segment that displays a summary of
performance of the trainee with respect to the simulated customer
interaction.
11. The computer readable medium according to claim 8, in which the
dialog code segment provides the predetermined customer portion by
playing at least a portion of an audio file.
12. The computer readable medium according to claim 11, in which
the response code segment receives the response to the customer
portion by receiving and storing a verbal response from the
trainee.
13. The computer readable medium according to claim 8, further
comprising: a recording code segment that records the response as
an audio file; and a playback code segment that plays the audio
file in response to a request.
14. The computer readable medium according to claim 8, further
comprising: a coaching code segment that provides instruction to
the trainee based on at least the response to the customer
portion.
15. The interactive training method according to claim 14, in which
the coaching code segment comprises a determining code segment that
determines a level of instruction to provide to the trainee prior
to providing the instruction, the level of instruction being based
on at least a determination of performance by the trainee.
16. A computer readable medium for storing a computer program that
provides interactive training, the computer readable medium
comprising: a retrieving code segment that retrieves a previously
stored, simulated customer interaction from a database, the
simulated customer interaction comprising a plurality of dialog
segments, each dialog segment comprising a customer portion and at
least one keyword corresponding to the customer portion; an
executing code segment that executes the plurality of dialog
segments by providing the customer portion of each dialog segment
to the trainee via a trainee terminal and receiving a corresponding
trainee response; and a determining code segment that determines
whether each trainee response includes the at least one keyword
corresponding to the customer portion of each dialog segment.
17. The computer readable medium according to claim 16, further
comprising: a coaching code segment that provides a predetermined
level of instruction to the trainee when the determining code
segment identifies an incorrect trainee response of at least one
dialog segment that does not include the at least one keyword
corresponding to the customer portion of the dialog segment.
18. The computer readable medium according to claim 17, in which
the coaching code segment further comprises a determining code
segment that determines a level of instruction based on at least a
number of times the trainee performed the simulated customer
interaction.
19. The computer readable medium according to claim 16, further
comprising: a display code segment that displays a summary of the
trainee's performance with respect to the simulated customer
interaction at the trainee terminal, the summary comprising a
number of trainee responses associated with the plurality of dialog
segments that included the at least one keyword corresponding to
each dialog segment.
20. The computer readable medium according to claim 16, further
comprising: an editing code segment that receives a change to a
selected dialog segment of the plurality of dialog segments, the
change comprising at least one of adding or deleting the at least
one keyword.
Description
[0001] This application is a continuation of pending U.S. patent
application Ser. No. 10/078,429, filed Feb. 21, 2002, the
disclosure of which is expressly incorporated herein by reference
in its entirety.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates to the field of interactive
computer-based training. More particularly, the present invention
relates to training call center agents at remote terminals
networked to a database, based on verbal and/or textual interaction
with simulated customers in dynamic scenarios.
[0004] 2. Background Information
[0005] Customer service agents working in call centers are
typically required to learn and internalize significant amounts of
information, including various policies, procedures, products and
systems, in order to competently perform their duties. In addition
to the large amount of information, many modern service industries,
such as the telecommunications industry, involve a high rate of
change as a result of evolving technology, regulations and consumer
appeal, further increasing the difficulty for the customer service
agents to remain current. Moreover, customer service call centers
historically suffer high turnover of personnel. The combination of
these factors requires the. expenditure of significant resources to
continually train customer service agents to reach and maintain
acceptable levels of performance.
[0006] The conventional approach to training call center agents
includes providing the necessary substantive information regarding
products, services, corporate policies and the like, teaching the
appropriate customer communication techniques and implementing
practice sessions to replicate (to the extent possible) actual
conversations with customers. The typical practice sessions are
conducted by trainer/instructor teams, in person. The training may
include standard policies and parameters, but these are not
consistently enforced or measured in an automated fashion. The
consistency of training and measure of progress, from one trainee
(or group of trainees) to another, is largely unknown. To simulate
realistic customer interactions, many call center training
operations use face-to-face role playing, which involves fellow
trainees or trainers acting as customers. The interactions, as well
as the accompanying feedback, are therefore inconsistent in quality
and content. Such variance in training and feedback may result in
disparate training results. Furthermore, statistical tracking of
training results is not particularly useful because of the
potential for inconsistency in the underlying training
methodology.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] The present invention is further described in the detailed
description that follows, by reference to the noted drawings by way
of non-limiting examples of embodiments of the present invention,
in which like reference numerals represent similar parts throughout
several views of the drawings, and in which:
[0008] FIG. 1 is a block diagram showing an exemplary training
network capable of enabling the interactive computer-based training
system, according to an aspect of the present invention;
[0009] FIG. 2 is a block diagram showing exemplary data and
applications for implementing the interactive computer-based
training system, according to an aspect of the present
invention;
[0010] FIG. 3 is a flowchart depicting an exemplary interaction
with a trainee using the interactive computer-based training
system, according to an aspect of the present invention;
[0011] FIG. 4 is an exemplary web page of the interactive
computer-based training system depicting a dialog segment of a
scenario, according to an aspect of the present invention;
[0012] FIG. 5 is an exemplary web page of the interactive
computer-based training system depicting trainee performance
results, according to an aspect of the present invention;
[0013] FIG. 6 is an exemplary web page of the interactive
computer-based training system depicting scenario registration,
according to an aspect of the present invention; and
[0014] FIG. 7 is an exemplary web page of the interactive
computer-based training system depicting coaching registration,
according to an aspect of the present invention.
DETAILED DESCRIPTION OF EMBODIMENTS
[0015] The present invention relates to an interactive
computer-based training (CBT) program directed to improving the
performance of personnel interacting with the public, especially
telephone-based customer service call center agents. The training
program enables the trainees to practice business conversations
with virtual customers in an "e-Learning" environment that provides
immediate remediation for errors, depth learning for tone and style
and the ability for trainers and managers to efficiently monitor
progress. The training program further enables training developers
to create scenarios, and trainers and managers to change scenarios
for the trainees to focus the training on specific areas, for
example, to address a trainee's demonstrated weaknesses, and to
implement the new or modified scenario in real-time.
[0016] The invention incorporates core CBT and web-based training
(WBT) standards, integrating leading-edge voice technology to
create realistic on-the-job experiences. In response to the dynamic
nature of typical telecommunications environments, and the need for
service representative training to keep pace, the training program
delivers a customizable, interactive learning experience and can be
controlled by trainers having no special level of technical skills
or computer programming knowledge. Also, the voice interaction is
speaker independent, such that the users do not have to "train" the
system prior to participating in the training.
[0017] An objective of the training program is to improve on-job
performance of call center agents. This requires standardizing
trainee interactions, capturing data regarding trainee behavior
relative to the objectives of specific business scenarios and
measuring gaps in experience. Vocal responses are solicited from
the trainees through speech recognition technology and measured for
analysis. A further objective of the training program is to enable
reporting, statistical analysis, and other administrative
capabilities to enhance management of customer service
representative training.
[0018] An aspect of the present invention provides a method for
interactively training call center personnel using a computer-based
system, the training including at least one call scenario involving
simulated customer interactions. The call scenario includes
multiple dialog segments, each of which includes a preprogrammed
customer portion and at least one keyword. The dialog segments are
sequentially executed by providing the customer portion of each
dialog segment to a terminal of a trainee and receiving a trainee
response to the customer portion of each dialog segment. It is
determined whether the trainee response includes the keywords
corresponding to each dialog segment. The trainee response
associated with each of the dialog segments may be recorded as a
sound file, which may be played in response to a request received
from the trainee terminal.
[0019] When each dialog segment has been sequentially executed, a
summary of the trainee's performance with respect to the call
scenarios may be displayed at the trainee terminal. The summary
includes the number of trainee responses that included the keywords
corresponding to each dialog segment. Also, the keyword associated
with a selected dialog segment may be replaced with an updated
keyword, which is different from the replaced keyword. It is then
determined whether the trainee response includes the updated
keyword.
[0020] Another aspect of the present invention provides a method
for providing computer-based, interactive call center training that
includes storing at least one call scenario in a database. Each
call scenario includes multiple dialog segments, which include a
customer portion and at least one keyword. A selection of a call
scenario is received through a trainee terminal. A customer
interaction is simulated between a trainee at the trainee terminal
and a simulated customer, according to the call scenario. The
simulation includes sequentially executing the dialog segments by
providing the customer portion of each dialog segment to the
trainee through the trainee terminal and receiving a corresponding
text and/or voice trainee response. The dialog segments may be
sequentially executed only when the trainee response includes the
at least one keyword corresponding to each dialog segment. It is
determined whether the trainee response includes the at least one
keyword. A selected dialog segment of the call scenario may be
edited on-line, which includes adding, deleting or changing at
least one of the customer portion and keywords corresponding to the
dialog segment.
[0021] Another aspect of the present invention provides a method
for interactively training call center personnel using an automated
customer interaction between a trainee and a simulated customer.
The method includes retrieving a call scenario from a database,
including at least one dialog segment, and providing a customer
portion of each dialog segment by displaying text or generating
speech at a trainee terminal. A trainee portion of each dialog
segment is received from the trainee terminal and includes at least
one of text and voice entered by the trainee. It is determined
whether the trainee portion of the dialog segment includes a
predetermined keyword associated with the dialog segment. When the
trainee portion does not include the predetermined keyword, a
predetermined level of instruction is provided to the trainee
relating to the dialog segment. The predetermined level of
instruction is set according to a skill level of the trainee.
[0022] When the trainee portion includes the predetermined keyword,
it may be determined whether the scenario includes at least a
second dialog segment. When the second dialog segment exists, a
customer portion of the second dialog segment is provided to the
trainee terminal by at least one of displaying text and generating
speech at the trainee terminal. A trainee portion of the second
dialog segment is received, which is at least one of text and
voice. It is then determined whether the trainee portion of the
second dialog segment includes a second predetermined keyword
associated with the second dialog segment. When the trainee portion
does not include the second predetermined keyword, the
predetermined level of instruction is provided to the trainee
relating to the second dialog segment.
[0023] With respect to setting the predetermined level of
instruction, the skill level of the trainee may be determined
automatically based on the number of times the trainee has
performed the call scenario. Also, the skill level of the trainee
may be determined based on input by a trainer setting the skill
level with respect to the trainee and the call scenario.
Determining the skill level may occur during the interaction
between the trainee and the simulated customer.
[0024] The results of the interaction between the trainee and the
simulated customer may be stored in a database. The results include
an indication of whether the trainee portion of the dialog included
the predetermined keyword and an indication of the level of
instruction provided to the trainee, if any. The results of the
interaction between the trainee and the simulated customer may be
compared with results of at least a second interaction between a
second trainee and the simulated customer. A report is generated
analyzing the comparison.
[0025] Another aspect of the present invention provides a computer
readable medium for storing a computer program that interactively
trains call center personnel. The computer readable medium includes
a memory segment that stores at least one call scenario, including
multiple dialog segments, and a training source code segment that
enables selection of a call scenario. Each dialog segment includes
a customer portion and at least one keyword. The computer readable
medium further includes a simulator source code segment, which
simulates an interaction between a trainee and a simulated customer
according to the call scenario, and an administration source code
segment, which enables administrative functions with respect to the
computer program. The simulator source code segment includes a
business rules source code segment that sequentially executes the
dialog segments by providing the customer portion corresponding to
each dialog segment to the trainee and receiving a corresponding
trainee response. The trainee response is text and/or voice entered
by the trainee. The administration source code segment includes a
call scenario editor source code segment that enables an addition,
deletion and/or edit of the call scenario.
[0026] The simulator source code segment may further include a
coach source code segment that determines whether the trainee
portion of each dialog segment includes the at least one keyword
corresponding to each dialog segment. The business rules source
code segment sequentially executes another dialog segment when the
keyword is included. The coach source code segment provides a
predetermined level of instruction to the trainee when the keyword
is not included.
[0027] The administration source code segment may further include a
reporter source code segment that generates a progress report of
the trainee. The progress report includes a performance assessment
based on the trainee including the keywords in the trainee
responses corresponding to the dialog segments and a comparison of
statistical data stored in the memory segment. The simulator source
code segment may further include a software emulation source code
segment that provides an emulation of actual call center software
to the trainee through a graphical user interface (GUI). The
simulator source code segment may also include a voice playback
source code segment that records the voice entered trainee portion
corresponding to each dialog segment and replays the recorded
trainee portion upon instruction received from a GUI.
[0028] Yet another aspect of the present invention provides a
computer readable medium for storing a computer program that
interactively trains call center personnel using an automated
customer interaction between a trainee and a simulated customer.
The computer readable medium includes a retrieving source code
segment, a simulator source code segment and a coaching source code
segment. The retrieving source code segment retrieves a call
scenario from a memory, the call scenario including at least one
dialog segment. The simulator source code segment executes the
dialog segments by providing a customer portion of each dialog
segment to a trainee at a trainee terminal and receiving a trainee
portion of the dialog segment. The coaching source code segment
determines whether the trainee portion of the dialog segment
includes a predetermined keyword associated with the dialog
segment. The coaching source code segment also provides a
predetermined level of instruction to the trainee relating to the
dialog segment when the trainee portion does not include the
predetermined keyword. The predetermined level of instruction is
set according to a skill level of the trainee.
[0029] Furthermore, the simulator source code segment may determine
whether the scenario includes at least a second dialog segment when
the trainee portion includes the predetermined keyword. When a
second dialog segment exists, the simulator source code segment
provides a customer portion of the second dialog segment to the
trainee at the trainee terminal and receives a trainee portion of
the second dialog segment. The coaching source code segment then
determines whether the trainee portion of the second dialog segment
includes a second predetermined keyword associated with the second
dialog segment. When the trainee portion of the second dialog
segment does not include the second predetermined keyword, the
coaching source code segment provides the predetermined level of
instruction to the trainee relating to the second dialog
segment.
[0030] The coaching source code segment may determine the skill
level of the trainee automatically based on a number of times the
trainee has performed the call scenario. The coaching source code
segment may determine the skill level of the trainee based on an
input by a trainer setting the skill level with respect to the
trainee and the call scenario. Determining the skill level may
occur during the interaction between the trainee and the simulated
customer.
[0031] The computer readable medium may further include a storing
source code segment that stores results of the interaction between
the trainee and the simulated customer in the memory. The results
include an indication of whether the trainee portion of the dialog
segment included the predetermined keyword and an indication of the
level of instruction provided to the trainee, if any. Also, the
computer readable medium may include a comparing source code
segment and a reporting source code segment. The comparing source
code segment compares the results of the interaction between the
trainee and the simulated customer with results of at least a
second interaction between a second trainee and the simulated
customer. The reporting source code segment generates a report
analyzing the comparison.
[0032] FIG. 1 is a block diagram depicting an exemplary data
network enabling the present invention. The network is modular,
which enhances flexibility to incorporate emerging technologies,
such as advances in voice recognition and simulation, without
having to redesign or replace the entire network. The exemplary
network of FIG. 1 includes a database 112, an associated database
server 110 and a web server 115, accessible through a packet
switched data network, such as Intranet 120. In an alternative
embodiment, the database server 110, the database 112 and the web
server 115 are accessible via secure connections through the
Internet using, for example, secure sockets layer (SSL) protocol,
developed by Netscape Communications.
[0033] The web server 115 may be an Internet Information Server
(IIS), developed by Microsoft Corporation, for example, running
Microsoft IIS 4.0 software. The database 112 may be a standard
relational database, for example, serviced by the database server
110, such as a Microsoft structured query language (SQL) server 7.0
running Microsoft SQL server 7.0 software. In an embodiment of the
invention, the database 112 has self-administering tools and
collects and stores data in tables. Each table represents a class
of objects representing features of the invention, such as
scenarios, dialog segments and keywords, as discussed below. The
application database 112 may be accessible by Microsoft's open
database connectivity (ODBC) and object linking and embedding
database (OLE DB) interfaces. Of course, any compatible databases
and servers having sufficient capability to implement the invention
may be incorporated.
[0034] The training program involves three categories of users: (i)
trainees, (ii) trainers and (iii) managers. In an embodiment of the
invention, each of the user categories accesses the training
program from a remote computer terminal or GUI, e.g., the trainee
terminal 130, the trainer terminal 150 and the manager terminal
170, networked with the web server 115 and the database 112 through
the intranet 120. The remote computer terminals may be, for
example, IBM Pentium based personal computers (PCs), running the
Microsoft Windows 98 or Windows NT operating systems, with standard
sound cards and Microsoft Visual Basic 6.0. Each PC further
includes a web browser, such as Microsoft Internet Explorer or
Netscape Navigator, available from Netscape Communications
Corporation. Alternative embodiments of the invention include any
number and combination of trainee, trainer and manager terminals
connectable to the intranet 120, including, for example, users from
all three categories accessing the training program from a single
terminal or any user category having multiple terminals configured
to access the training program simultaneously.
[0035] The trainees interact with the training program from the
trainee terminal 130 by voice or text in order to train and
practice call simulations. The voice interaction is enabled by a
headset with a microphone, such as a Plantronics Model LS1 stereo
PC headset, connected to the trainee terminal 130. The trainee
responds to simulated customer input by speaking into the
microphone or typing on a keyboard connected to the computer
terminal.
[0036] The trainers and managers communicate with the training
program primarily by text interaction from the trainer terminal 150
and the manager terminal 170, respectively. The trainers manage the
training simulations, coach the trainees and report on the various
scenarios. The trainers are able to alter scenarios to particularly
address areas of weakness demonstrated by a trainee. Altering the
scenarios may be performed in real-time, when desired by the
trainer. The trainers are also able to adjust the level of
automated feedback available to the trainees, also known as
coaching.
[0037] The managers maintain the training program, create and
control simulations, and manage access to the training program.
Depending on the implementation of the training program, the
managers may be divided into administrators and developers. The
administrators perform purely administrative functions, such as
maintaining the training program and data, controlling access to
the system, generating and analyzing reports, and monitoring the
overall performance of the program, as well as the trainees. The
specific functionality of each administrative function is discussed
in detail, below. The developers, who may also be administrators,
focus on the content of the training program, which includes
writing and updating scenarios, as well as the scope and substance
of the automated feedback. Like the trainers, the developers are
able to alter scenarios to address areas of importance or concern
for the training program to particularly address.
[0038] The interactive training program software is modular. In
other words, the training content is segmented among various call
scenarios, along with the supporting knowledge and skill bases. The
modular nature of the training program has several advantages. For
example, the training scenarios may be assembled to fit specific
short-term and long-term training targets. Also, in the event of a
change in content, such as a change in policy or technology of the
calling center for which the trainees are training, the affected
scenarios may be easily isolated and modified, as needed. The
modular design is scalable and flexible, allowing parallel
development of the various modules, as well as customization and
iteration of the overlying interfaces, without affecting the
underlying training program logic. The modular design enables the
training program to efficiently adjust to business needs caused by
rapid and dramatic changes within the relevant industry.
[0039] In an embodiment of the invention, the interactive training
program includes four basic modules, each including various
applications and/or dynamic data, as depicted in the block diagram
of FIG. 2. The modules include the database 212, the training
module 220, the simulator module 230 and the administration module
240, each of which includes multiple applications related to the
various features of the present invention. The database 212 is
accessible to the other modules through an interface, e.g.,
application programming interfaces (APIs) 214, enabling the
applications access to the associated data. The APIs provide
consistent, yet flexible interfaces that enable changes or
additions to be made within one module without requiring
corresponding changes in the database 112 or the other modules. In
an embodiment of the invention, the applications are correlated
with objects implemented in object oriented programming.
[0040] The processing of the training program may be centralized,
such that each module is stored and implemented by a central server
or other processing unit. Alternatively, the training program may
be distributed across a combination of integrated processors. For
example, the database 212 may be located in a centralized database,
such as the database 112 of FIG. 1, accessible by numerous users
over a broad geographic area through a packet switched data
network, such as the intranet 120. The training module 220 and the
administration module 240 may be web based applications. The engine
that drives the training module 220 and/or the administration
module 240 may reside on the database server 110 or the web server
115, having logic implemented, using Java or JavaScript. The
simulator module 230 may be a client based application executed,
for example, by the trainee terminal 130. Regardless of the
distribution of the processing, the functionality of the database
212 and each application module remains essentially the same.
[0041] The database 212 interfaces with all modules of the training
program in order to drive the respective application with dynamic
data. The data stored in the database 212 includes the various
scenarios created to provide the trainee with a realistic, yet
controlled interaction with a simulated caller. Each scenario
includes a series of dialog segments, which are the sets of
customer lines or inputs, and the corresponding trainee responses.
The dialog segments are designed to represent categorical customer
inputs that a call center agent would expect to hear on the job at
an actual call center. For example, a scenario may be designed
around a sophisticated customer, an angry customer, a
non-communicative customer, a confused customer, a new customer and
the like.
[0042] Each dialog segment includes a set of keywords. The keywords
are the words determined by the trainers and/or managers to be
critical to the customer interaction and must be spoken by the
trainee in order to progress within the scenario, as determined by
the call flow business rules 232 and the coach application 234,
discussed below. In an embodiment of the invention, the keywords
are classified into multiple levels, such as mandatory words, which
must be spoken (or typed) and are crucial to acceptable performance
under the training scenario; forbidden words, which cannot be used
by the trainee under the circumstances of the scenario; and
optional words, which are recognized alternatives to the mandatory
words, but are scored separately. Each type of keyword also has
respective aliases, which are essentially synonyms of keywords. For
example, aliases of the mandatory keywords may be identified at the
time a scenario is created and stored along with the associated
mandatory keywords in the database 212. The aliases are
interchangeable with the mandatory keywords from a scoring
perspective. Also, because of the flexibility of the training
program, the trainers may add aliases words through the scenario
editor 242 of the administration module 240, discussed below, at
any time. For example, when it becomes apparent that another word
is used frequently and is just as effective as the associated
mandatory keyword, the trainer may decide to include the
alternative word as an acceptable alias.
[0043] The keywords may be required for each dialog segment of the
scenario or, more generally, for the overall scenario. In other
words, depending on the design of the training program, the trainee
may receive credit for including keywords in the conversation when
the keyword is said during a particular dialog segment in which the
use (or non-use) of the keyword is determined to be most critical.
In a less stringent version, the trainee may receive credit for
keywords used anytime during the scenario. The keywords also
include criteria levels, which enable the simulator module 230 to
identify the success level of each spoken dialog. The managers may
classify the importance levels of each keyword while building the
scenarios. The importance levels may likewise be updated through
the scenario editor 242.
[0044] The simulator module 230 simulates a typical telephone call
to a customer service call center to solicit responses from the
trainee. The simulator module 230 contains the training logic and
controls the GUI that the trainee uses to practice a simulation.
The simulator module 230 may be a client-server application, for
example, residing on the individual workstations, such as the
trainee terminal 130. The trainee initiates the simulator module
230 from the training module 220, which may be web-based, by
selecting a particular training scenario. In response, the
simulator module 230 loads simulation information from the database
212, including keywords, dialog segments and sound files.
[0045] The simulator module 230 creates a realistic training
environment in two respects. First, through the business rules
application 232, the simulator module 230 initiates a real-time,
spoken conversation with the trainee at a GUI, e.g., trainee
terminal 130, incorporating the pre-established dialog segments
from the selected scenario. Second, the software emulation 238 of
the simulator module 230 is programmed to mimic the software
actually running a customer service center for which the trainee is
training. For example, the screenshots, caller data format,
options, keyboard configuration and the like are identical to that
of the actual call center, so the trainee is trained on system
operation at the same time the trainee is learning to communicate
with customers.
[0046] The business rules application 232 also identifies data to
be loaded from and stored to the database 212, as well as starts
and stops a voice recognizer, which may be client-based on the
trainee terminal 130, for example, and which is integrated with the
simulator module 230. The voice recognizer listens to human speech
and extracts keywords provided by the business rules application
232. When appropriate, according to the pre-established rules
governing the training session, a computerized coach application
234 is activated to assess the trainee's actions and voice
responses based on the keywords associated with the dialog
segments, which are uploaded to the simulator module 230 from the
database 212, in an embodiment of the invention. Upon completion of
the selected scenario, the business rules application 232 also
determines and displays the performance statistics or score at the
trainee terminal 130, enabling immediate self-assessment by the
trainee.
[0047] The coach application 234 is a virtual instructor. It
receives the trainee's statements by way of the voice recognizer
and provides measured feedback, such as hints, tips, learning units
and other assistance based on the trainee's success relative to the
selected level of coaching. The coach application 234 monitors the
trainee's performance based on keywords, evaluates the use of each
keyword and provides feedback to the trainee regarding the level of
performance at the end of each dialog segment or scenario. The
coach application 234 also provides the trainee the opportunity to
type responses to computer generated customer input, in the event
the trainee feels (or the business rules application 232
determines) that the voice recognition was inaccurate. When
performance is unacceptable, the coach application 234 may provide
feedback and require the trainee to repeat the particular dialog,
as discussed below. Furthermore, in an embodiment of the invention,
the coach application 234 may enable performance support tools
external to the training program. For example, the coach
application 234 may hyperlink to another training system or data
feed more specifically directed to a shortcoming repeatedly
demonstrated by the trainee.
[0048] The simulator module 230 also includes software emulation
238 that includes interactive screen shots designed to represent
actual call center desktop software. Because different companies,
or different regions within the same company, may vary with respect
to call center design and implementation, the software emulation
238 may include multiple screen shot emulations representing the
software of the respective regional call centers. The software
emulation 238 may reside in a regional database, for example. The
software emulation 238 duplicates regional software behavior to the
degree required by a particular call simulation or training goal,
and thus provides functionality needed to complete the scenario in
a realistic manner suitable to the performance and learning
need.
[0049] In an embodiment of the invention, the software emulation
238 is enabled by corresponding, region-specific data, which may
likewise be stored in a regional database and updated through a
network (e.g., intranet 120) interface, enabling efficient
modification. For example, when a region changes an offered product
or service, the call service representatives must be retrained.
Data describing the new product or service is stored in the
regional database, replacing data describing the old product or
service. The software emulation 238 for the affected region may
then proceed as always, drawing on the newly stored regional data,
as represented in simulated screens mimicking the new desktop
application supporting the new product.
[0050] The regional database may also include customer data, which
likewise populates the software emulation 238, so that the trainee
receives "pop-up" caller data during simulated calls, as would be
the case in an actual call center interaction. The customer data
includes, for example, the customer name, address, account balance,
current product/service information and the like. The customer data
may also be updated by a web interface.
[0051] The various users interact with the training system by way
of the training module 220. For example, the trainees access the
training portion of the training program through the training
module 220, while the trainers and managers likewise interface with
both the training portion of the training program and the
administration module 240 through the training module 220. For the
trainee, the training module 220 displays a home page on a GUI,
such as the trainee terminal 130, enabling the trainee to login,
customize the interface, view and select authorized training
scenarios, and the like. The training module 220 includes, for
example, a scenario catalog 222, a user statistics application 224
and a profile editor 226, discussed below.
[0052] A trainee enters the training program by logging onto the
system from a home page displayed at the trainee terminal 130 at
step s310 of FIG. 3. The log-in process is enabled by pre-assigned
identification data specific to the trainee, such as an account
number. (or user name) and a password. The account number and
password are determined by the manager and assigned to the trainee
at the beginning of the training process. Data specific to the
particular trainee may also be stored in association with the
account number, such as the experience level of the trainee (e.g.,
new training, refresher course, continuing education on new
systems, etc.). The log-in is performed in response to a typical
introductory screen and is entered through the terminal keyboard
or, in an alternative embodiment, through voice commands. The
training system authenticates the user by retrieving the account
number and password from the corresponding file in the database 212
and determining the validity of the same, as well as determining
and implementing any restrictions to system access that may be
associated with the trainee.
[0053] After allowing the trainee to log-in, the simulator module
230 causes options to be provided to the training terminal 130 at
step s312. The options may be provided textually or audibly. The
options include viewing scenarios, statistics or profile data,
changing the log-in data associated with the account number, and
logging out. As shown in FIG. 2, the scenarios are viewed (and
ultimately executed) by way of a scenario catalog 222, which may
include scenario catalogs directed to the specific trainee or to
the class or region to which the trainee belongs. Viewing
statistical data is enabled through a user statistics application
224 and viewing (and editing) profiles is enabled through a profile
editor 226. The initial log-in screenshot may include additional
information relating to the trainee, such as the trainee's name,
the associated trainer's name, the number of scenarios the trainee
completed, the number of scenarios for which the trainee is
registered and the like.
[0054] When the trainee elects to view the scenarios at step s314,
a list of scenarios is displayed on the screen of the trainee
terminal 130. The scenarios include practice scenarios, which may
be performed by the trainee at will to hone skills without
affecting the trainee's overall score. Performance of the practice
scenarios may be limited, depending on the parameters established
by the trainers or managers. Similarly, the trainee may select a
registered scenario, the availability of which is predetermined by
the trainers or managers. Unlike the practice scenarios, the
results of the registered scenarios are stored for performance
evaluation. The scenario display may include the name and a brief
description of the scenario, together with an indication of whether
the scenario is a practice scenario or a registered scenario. To
select a training scenario, the user clicks on the designated
portion of the display, such as specific text or an icon,
representing the desired scenario, which activates the simulator
module 230.
[0055] In an embodiment of the invention, the trainee accesses a
catalog of scenarios from the scenario catalog 222 and specifically
requests permission to enter and execute scenarios the trainee
deems helpful. An example scenario catalog 222 is depicted in box
410 of web page 400 shown in FIG. 4. Each scenario may include
parenthetical information to guide the trainee, such as a
description of its purpose or the state or service to which the
scenario applies.
[0056] At step s315, the scenario selection is received from the
trainee terminal 130 and authorized at step s320. As stated above,
the trainers and/or managers authorize the practice and registered
scenarios that a trainee (or group of trainees) is permitted to
execute. The access information is created and updated through the
registration system 244 and stored in the database 212. In an
embodiment of the invention, only the scenarios for which a trainee
is pre-approved are displayed to the trainee in response to step
s314, eliminating the need for authorization at step s320.
[0057] When the selected scenario is authorized, the simulator
module 230 initiates the scenario at step s322. In an embodiment of
the invention, the data relating to the scenario, including the
audio files and/or voice generation files for customer simulation,
are uploaded from the database 212 to the simulator 230. The
relevant dialog segments may then be stored on a local hard drive,
for example, of the trainee terminal 130, or alternatively in some
other virtual digital resource.
[0058] With respect to the voice interaction, the trainee receives
prerecorded audio files or, alternatively, audio signals from a
voice simulator interfaced with the database 212, which simulate a
conversation from the customer's perspective. For example, extended
mark-up language (XML) strings of text may be received via a voice
simulator API in the APIs 214 using hypertext transfer language
protocol (HTTP). In an embodiment of the invention, the voice
simulator is based on RealSpeak for Windows text-to-speech
software, developed by Lernout & Hauspie Speech Products USA,
Inc., which is capable of providing about 16 hours of speech for
every one hour of system operation.
[0059] The received data specifies a phrase to be spoken by the
computer, a unique identification number associated with the phrase
and various control parameters, such as speech rate and volume.
Upon receipt of the text data, the voice simulator API responds by
acknowledging the request and storing the phrase in a processing
queue table. At pre-determined time intervals, a voice simulator
operating with the simulator module 230 checks the processing queue
table for new entries. When a new entry is identified, the voice
simulator sends the phrase to a text-to-speech application to
transform the text data into audio files, which are stored, for
example, in .WAV format, developed by Microsoft and IBM, or MP3
format. The audio files may be stored at the web sever 115 and thus
retrievable by a universal resource locator (URL) or universal
naming convention (UNC) path.
[0060] Each audio file is assigned a unique name based on a dialog
segment identification in a dialog table. Once the audio file is
processed, either successfully or unsuccessfully, the status code
of the dialog table is updated. The status code, together with the
file name, is provided to the scenario editor 242 in order to show
the trainers or managers which audio files have been completed. The
voice simulator API may enable deletion of selected audio files, as
well as responses to requests for default values and available
parameter values. More than one voice simulator may be implemented,
depending on the system capacity to deliver the desired number of
hours of speech in one hour.
[0061] In an embodiment of the invention, sound files may be
generated using sources other than the voice simulator. For
example, actual voice recordings may be made and stored as .WAV or
MP3 files. Like the generated voice files, each sound file is
assigned a name and location and stored in the web server 115. The
stored voice strings and associated file names and locations are
passed to the simulator module 230 for implementation.
[0062] In response to the scenario initiation, the business rules
application 232 receives input from the trainee at step s324. The
trainee's input includes speaking into the headset microphone or,
alternatively, typing on a keyboard interfaced with the trainee
terminal 130. Note that the initial dialog segment of the scenario
includes an incoming customer call and the trainee's response,
e.g., answering the call and greeting the customer. For example,
the software emulation 238 may provide an indication of a queued
call that mimics the same indication received by an agent at the
actual call center, such as a beeping call initiation sound,
indicating that the scenario is ready. The trainee receives the
call (e.g., by a particular keystroke, as provided by the software
emulation 238) and responds with a greeting.
[0063] In an embodiment of the invention, the current dialog
segment is displayed on the screen contemporaneously with execution
of the trainee's portion of the dialog segment. For example, in the
web page 400 of FIG. 4, the dialog counter 411 indicates "1," i.e.,
the first dialog segment. A short description of the dialog segment
number 1 is indicated in the call flow block 412 as "greet the
customer." The entire dialog segment is captured as a customer
statement 413 followed by the trainee response 414. Because FIG. 4
depicts the initial trainee interaction, the customer statement 413
merely states "indication of incoming call," eliciting the trainee
response 414: "Thank you for calling Ameritech. My name is Adam.
How can I provide you with excellent service today?"
[0064] The voice recognizer integrated with the simulator module
230 receives the trainee's voice input. In an embodiment of the
invention, the voice recognizer is implemented as a Microsoft
ActiveX control compatible with Microsoft Visual Basic, Version
6.0, as a component object model layer on top of ASR 1600 software,
for example, developed by Lernout & Hauspie. Furthermore, the
voice recognizer is preferably speaker independent. In other words,
it does not need to be "trained" based on each individual user's
voice. The voice recognizer listens to human speech captured, for
example, at the trainee terminal 130 by a headset microphone.
[0065] The spoken responses are converted into text by the voice
recognizer, which text is compared through the coach application
234 to the pre-established keywords retrieved from the database 212
relating to the current dialog segment at step s326. The original
keywords are received, for example, via a keyword API over HTTP
having XML format. At step s330, the coach application 234
determines whether coaching is to be provided to the trainee. As
stated above, the coach application 234 is programmed to provide
hints, tips and learning units in direct correlation with the
trainee's performance, and displays the performance statistics or
score at the trainee terminal 130. The determination of the coach
application 234 is two-fold. The coach application 234 determines
the trainee's performance results (e.g., the number of keywords
successfully included, or not included, in the trainee's input),
and whether the performance results warrant coaching (remediation)
in view of the intervention level set, for example, by the
trainers.
[0066] The intervention level is the threshold that governs the
amount of coaching to be provided to a particular trainee or class
of trainees. A new trainee may warrant a high level of coaching,
while a trainee nearing the end of the training program may warrant
little to no coaching. For example, when the intervention level of
coaching is set to fifty percent, the coach application 234
interrupts the scenario between dialogs and provides guidance
whenever the trainee fails to match half of the required keywords.
Similarly, an intervention level of zero percent means that the
coach application 234 does not interrupt the scenario at all,
regardless of the trainee's performance. The intervention level is
set by the trainer through the administration module 240. However,
in an embodiment of the invention, the trainer may adjust the
intervention level through the training module 220 for individual
trainees at their terminals.
[0067] When the coach application 234 determines that coaching is
required, it provides substantive feedback at step s332. The
content of the substantive feedback may likewise be set to
correspond to the significance of the errors or the skill of the
trainee, prior to or during the training session. In particular,
the coach application 234 may be programmed to provide a hint
whenever the trainee omits a keyword while responding to the call
initiation or the customer input of the current dialog segment. For
example, as shown in the hint block 420 of FIG. 4, when the trainee
neglects to include his or her name in the greeting, the coach
application 234 provides a hint relevant to the omission without
actually giving the answer, e.g., "You need to greet the customer
and introduce yourself."
[0068] Similarly, the coach application 234 may provide a tip based
on additional predetermined criteria, such as when the trainee has
repeated a mistake. The coach application 234 may be programmed to
provide tips based on other criteria, as well, such as inexperience
of the trainee or participation in a practice scenario (as opposed
to a registered scenario). A tip is more definitive than a hint and
provides limited substantive guidance. The tip block 422 of FIG. 4
provides such additional guidance to the trainee, e.g., "Make sure
you identify your company; make sure you tell the caller your name;
make sure you offer to provide excellent service."
[0069] The coach application 234 may also provide the actual
keyword for which it was looking in the dialog segment, as
indicated for example in the keyword block 416 of FIG. 4. In an
embodiment of the invention, the keyword block 416 includes
mandatory keywords 418 and associated aliases 419 to further assist
the trainee. As discussed above, aliases are determined by the
trainers and/or managers prior to scenario execution in order to
provide the trainee controlled flexibility. For example, the
keyword block 416 informs the trainee that "assist" and "help" are
acceptable substitutes for the keyword "service." The trainers
and/or managers create and implement the aliases through the
scenario editor 242. Due to the ease with which the scenarios are
edited, a trainer may add aliases via the scenario editor 242 and
reissue the scenario for use, when the trainer determines that
particular words are being used repeatedly and appear to be equally
acceptable in the context of the scenario.
[0070] The highest substantive coaching level is provided in the
form of learning units. Learning units are displayed to the trainee
whenever more significant or repetitive errors occur during a
dialog segment or scenario. They provide more than simply
identifying a particular omission or mistake. Learning units are
actual lessons, providing background and explanation regarding the
subject matter with which the trainee is having difficulty. For
example, the learning units may provide actual course materials to
facilitate further learning by the trainee. In an embodiment of the
invention, the learning units include prerecorded video files that
enhance the trainee's learning experience by visual explanation
and/or demonstration.
[0071] At step s334, the coach application 234, in conjunction with
the business rules application 232, determines whether the dialog
segment that has resulted in feedback to the trainee should be
repeated. In an embodiment of the invention, whether to repeat a
dialog segment is based on the type of substantive feedback
provided. For example, when the training program provides a hint to
the trainee, repeating the scenario is necessary so that the
trainee has an opportunity to correct an error with minimal
coaching intervention. Providing a learning unit, however, results
in the trainee proceeding to the next dialogue segment because it
is likely the trainee will be required to repeat the entire
scenario, regardless. Alternatively, though, the dialog segment may
be repeated following a learning unit in order to underscore the
subject matter in the practical application. When the dialog
segment is repeated, the training program returns to step s324 to
receive the trainee's input. When the segment is not to be
repeated, the training program advances to step s340.
[0072] In another embodiment of the invention, the trainee is able
to navigate through each scenario, repeating the various dialog
segments as desired. The trainee may navigate, or be barred from
navigating, depending on how the trainer sets up the parameters of
the scenario. The trainer or manager enables the navigation ability
from the registration system 244, discussed below. The trainee may
then determine the problematic dialog segments, in conjunction with
the automated coaching, and repeat the dialog segments at will. The
trainee would not be able to navigate within registered scenarios,
which are scored for performance evaluation and therefore must be
standardized among the trainees.
[0073] At step s340, the business rules application 232 determines
whether the simulated call includes another customer line in the
scenario script. When there are no further customer lines, the
business rules application 232 determines that the scenario has
ended. The trainee's performance results are then displayed on the
trainee terminal 130, at step s344, for the trainee to assess his
or her own performance. The display of performance results may
operate independently from the coaching function. For example, when
a trainee's coaching level is set to zero, the trainee receives no
coaching during the scenario, but still receives performance
results upon completion of the scenario.
[0074] The performance results include, for example, an overall
score, the number of keywords in the scenario, the number of
keywords successfully included by the trainee, identification of
strengths and weaknesses, and the like. The performance results may
further include statistical analysis, such as comparisons of the
trainee's scenario performance against industry training standards,
company-wide or region-wide training standards, other trainees and
the like, which requires access to additional information stored in
the database 212.
[0075] In addition to the summary performance results displayed at
step s344, the trainee's performance progress may be tracked on the
client side and displayed during execution of the scenario, e.g.,
between dialog segments before being uploaded to the web-based
scoring application, in an embodiment of the invention. As shown in
exemplary web page 500 of FIG. 5, a performance progress block 501
is displayed along the left side of the trainee's screen throughout
the scenario interaction. A dialog table 502 indicates the number
of keywords recognized in each dialog segment. The dialog table 502
includes a vertical progress bar 504 showing the number of keywords
per dialog segment recognized by the voice recognizer and coach
application 234. The exemplary dialog table 502 provides for five
keywords per dialog segment. For a perfect scenario score, the
progress bar 504 for each of the dialog segments extends vertically
through all five keyword indicators.
[0076] Below the dialog table 502 is a dialog segment overview
section 512, which includes a dialog segment identifier 514 and
corresponding dialog segment data. The dialog segment identifier
514 indicates that the current dialog segment is the first of 10
dialog segments that comprise the scenario. The dialog segment data
includes the date and time 515 of the dialog segment, the score
516, the comparable average 517 of all trainees, the elapsed time
518 used to execute the dialog segment and the total coaching time
519 provided to the trainee. Additional alternative embodiments may
include any data that the trainers and/or managers deem appropriate
for display during the scenario.
[0077] Also, the trainee may hear his or her spoken response to a
customer's portion of the displayed dialog segment by clicking on
the speaker icon 520. The recording enables the trainee to hear the
actual verbal reply, including tone, volume, inflection and other
basic communication factors critical to the trainee's success as a
call center agent. The recordings may be stored locally at the
trainee terminal 130, for example, and accessed repeatedly to gauge
progress.
[0078] Referring again to FIG. 3, when the business rules
application 232 determines at step s340 that additional customer
lines are pending, the business rule application 232 instructs the
voice simulator to provide the next sequential customer portion of
the conversation at step s342. As described above, the customer
portion of the dialog segment is either a prerecorded voice or
computer generated text-to-speech statement. Alternatively, the
customer portion may be a textual statement.
[0079] The customer portion of the conversation initiates the next
consecutive dialog segment of the scenario, necessitating an
appropriate response from the trainee at step s324. Steps s324
through s344, as described above, are then repeated in relation to
the next dialog segment. In particular, the simulator module 230
receives the trainee's response through the voice recognizer at
step s324. The coach application 234 then identifies keywords at
step s326, determines whether coaching is needed at step s330,
provides feedback at step s332 (when required) and determines if
repeating the dialog is necessary at step s334 (when required). The
business rules application 232 then determines whether another
customer interaction is to be played at step s340 and provides the
next customer interaction at step s342, until the scenario is
completed.
[0080] Returning to step s314, when the viewing scenarios option is
not selected, the trainee may elect to activate the user statistics
application 224 at step s350. When the user statistics application
224 is selected, the corresponding options are displayed at the
trainee terminal 130 at step s352. Because the user statistics
application 224 is also available to the trainers and managers,
they may also be displayed at the trainer terminal 150 and the
manager terminal 170.
[0081] The user statistics application 224 provides training
simulation statistics at step s354, which may include the trainee's
scored performances for the executed scenarios, in the context of
the training requirements and goals, depending on the selection
made from the displayed options at step s352. The data is
standardized and may include additional information for the trainee
to conduct an informed evaluation of his or her performance in
comparison to other trainees in the trainee's group or throughout
the company. The data displayed by the user statistics application
224 is similar in content to the data displayed, for example, in
the exemplary web page 500 of FIG. 5, discussed above. However, the
data is formatted to incorporate multiple scenarios and
students.
[0082] The training program managers may customize the data
presentation and emphasize particular areas of performance. For
example, when the managers detect a trend toward excessive average
time spent with callers, either through evaluation of the training
program results or actual call center data, the user statistics
application 224 may be modified to highlight the total time per
call entry, and specifically note whether the time is too high. The
user statistics application 224 may similarly be customized by
trainers to focus the trainee (or group of trainees) on a
particular area of importance or weakness.
[0083] The trainee may alternatively elect to enter the profile
editor 226 at step s360. When the profile editor 226 is selected,
the authorized functionality for the trainee is displayed at step
s362, based on the authorized level of access pre-established by
the registration system 244 and/or the user manager 246. The
trainee's level of access is normally very limited. For example,
the trainee may access the pre-approved scenarios and user
statistics data, as well as change the trainee's password. In an
embodiment of the inventions the profile editor 226 may be used to
observe the various restrictions and privileges specific to the
trainee, as determined through the registration system 244,
discussed below.
[0084] Note that other user types are able to access and initiate
changes to trainee authorized functionality by way of the profile
editor 226. For example, a trainer is able to selectively block
trainee access to learning units. When a user is able to perform
edits, such as changing a password or altering authorized
functionality, the changes are received by the profile editor 226
at step s364. The database 212 is updated accordingly at step
s366.
[0085] The administration module 240 of the interactive training
program enables trainers and managers to retrieve and analyze
scenario, trainee, company and region-specific data, as well as to
edit existing scenarios. Furthermore, the managers are able to
control the operation, replication and distribution of the overall
training program, as well as add and delete scenarios (e.g., the
developers in particular). The administrator module 240 includes a
scenario editor 242, a registration system 244, a user manager 246
and a reporter system 248. In an embodiment of the invention, the
administrator module 240 may be a server application, for example,
residing on a network server, such as the web sever 115. The user
may initiate the administrator module 240 from the training module
220 by selecting the desired administrative function.
[0086] The scenario editor 242 provides a common interface enabling
modifications and maintenance required for each training scenario.
The scenario editor 242 may be a web-based, Java or JavaScript
application that generally enables upper level users, such as
trainers and managers, to read, edit and delete the scenarios
implemented by the business rules application 232, including the
dialog segments and the keywords. For example, existing objects may
be edited and new objects may be created.
[0087] Interaction with the scenarios must follow an appropriate
order. For example, the user must first select a scenario before
being able to view or edit a particular dialog within the scenario.
Similarly, the user must first select the dialog before being able
to view or edit the keywords associated with that dialog. The
customer data and the software emulation 238, which enable aspects
of the training program, may also be viewed and edited by way of
the scenario editor 242.
[0088] Also, the developers may add entirely new scenarios by way
of the scenario editor 242. In an embodiment of the invention, the
developer logs onto the training program system from the manager
terminal 170, which receives XML web pages via the intranet 120
from the web server 115. The web pages include fields for the
customer lines of multiple dialog segments. The developer types the
desired customer line and further provides the associated keywords
that need to be in the trainee's response. Alternatively, the
developer may reference voice files, prerecorded and stored or
produced on-line by the developer, for each of the desired customer
lines in a dialog segment. The data for the new scenario is then
stored in the database 212 as text. However, in an embodiment of
the invention, the customer lines are converted to speech (unless
they are already recorded speech) and also stored as audio files in
.WAV or MP3 format, for example.
[0089] In an embodiment of the invention, the scenario editor 242
likewise enables the trainers and managers to edit, read and delete
the content of the coaching application 234. The editing
functionality includes changing the substantive content of the
hints, tips and learning units administered via the coach
application 234.
[0090] The registration system 244 tracks and manages the
registration status of trainees to the various scenarios. The
functionality of the registration system 244 enables registering
and un-registering trainees for scenarios, as well as allowing and
listing all registered and practice scenarios for each trainee. The
trainee may be able to observe his or her registration status
through the profile editor 226, discussed above.
[0091] As shown in the exemplary web page 660 of FIG. 6, the
registration system 244 may display the registration status of the
trainees with respect to the available scenarios. In the depicted
embodiment of the invention, all of the scenarios are listed in the
scenario display 608 as headers 610 through 613. The web page is
configured to scroll horizontally to show the additional scenarios,
if needed. The individual trainees are listed in the associate
column 620.
[0092] Each trainee is specifically designated as registered or not
registered for each scenario. In an embodiment of the invention, a
scenario may be designated "all" or "none" with respect to each
registration, which relieves the trainer or manager from having to
individually indicate every trainee with respect to particular
scenarios. There is also an indication of whether practice
scenarios are permitted for each scenario. For example, in row 621,
representing Jane User, blocks 622 and 623, respectively, indicate
that she is registered for the first scenario with no practice
allowed, which means that the results of Jane User's performance of
the first scenario will be scored and stored in the database 212.
In the row 624, representing Kelly User, blocks 625 and 626,
respectively, indicate that she is registered for the first
scenario for purposes of practicing the scenario. In other words,
Kelly User's results will not be stored. A trainee may not be
registered for the same scenario more than once. This provides
control over the integrity of the training process and is
especially important for accurately comparing performance among
multiple trainees or groups of trainees.
[0093] The registration system 244 may also be used to customize
the nature of each trainee's interaction with the training program,
including setting the level of coaching, the level of software
emulation and the ability to navigate among dialog segments in
scenarios. Exemplary web page 700 of FIG. 7 contains an exemplary
individual trainee registration table. Entry 710 of the
registration table 700 is the mandatory keyword intervention level,
which indicates the percentage of mandatory keywords the trainee
must miss before the coach application 234 intervenes to provide
assistance in the form of hints, tips and learning units, as
discussed above. Entry 712 indicates that the mandatory
intervention level is set at 50, which means that coaching
intervention begins when the matched keyword count is less than or
equal to 50 percent. When entry 712 is set at 100, coaching
intervention begins when the matched keyword count is less than or
equal to 100 percent. When entry 712 is set at zero, there is not
coaching intervention regardless of keyword matches.
[0094] As discussed above, alternative embodiments of the invention
include additional types of keywords that the training program will
track and score. For example, entries 714 and 718 are the optional
keyword and forbidden keyword intervention levels, respectively.
The optional keyword intervention level 714 is the percentage of
optional keywords the trainee must match before receiving coaching
related to the use of optional key words. Note that the optional
keywords are recognized as being similar to the corresponding
mandatory keywords, indicating the trainee's basic understanding of
the scenario requirements, but are otherwise not acceptable for
purposes of passing the scenario. The forbidden keyword
intervention level 718 is the percentage of forbidden keywords the
trainee must use before receiving coaching related to the use of
forbidden key words. Both entries 716 and 720 indicate that the
optional keyword intervention level and the forbidden keyword
intervention level have been set at zero, meaning that the trainee
will receive the corresponding coaching whenever an optional
keyword or a forbidden keyword is used.
[0095] Entries 722 and 726 relate to whether software emulation 238
is to be explained and activated, respectively. Entry 724 indicates
by the checkmark that intervention explaining call center software
specific interactions is available to the trainee upon request.
Entry 728 indicates that the software emulation 238 is turned ON
for the trainee. When the software emulation 238 is OFF, the
trainee is presented the scenario on a generic software
application, which operates at a simplistic level with very limited
service and customer data options.
[0096] Entry 730 enables coaching to be turned OFF entirely. For
example, the trainer may want to test an advanced trainee, without
access to coaching, to get an accurate demonstration of the
trainee's unassisted performance. Entry 732 indicates that the
coaching application 234 has been enabled for the trainee. Entry
734 enables navigation by the trainee within the scenario. Again,
the trainer may want to eliminate the trainee's ability to move to
repeat dialog segments to observe a uninterrupted scenario
performance. Entry 736 indicates that the trainee is able to
navigate, if desired. The registration setting are stored in
conjunction with the trainee's scores for the registered scenarios.
The various settings will be reflected in the reports generated by
the reporter system 248, discussed below.
[0097] The user manager 246 of the administration module 240
enables users with administrative privileges to perform
administrative functions through a common interface. For example,
by way of the user manager 246, a manager is able to add new user
records, edit user information, remove user records and read user
information. Moreover, the manager may assign multiple levels of
permissive access to the training program, individually or by
category. For example, the manager determines the level of access
available to all trainers.
[0098] In an embodiment of the invention, the various levels
available to each user are as follows: Trainees (level 1) are able
to take training simulations for scores, view their statistical
information and performance results and change their passwords. The
trainees possess no administrative rights and cannot change
substantive data, including scenario registration, coaching
intervention levels and the like. Trainers (level 2) are able to
assign training scenarios to trainees, practice training
simulations and access the scenario editor 242, the registration
system 244 and the reporter system 248. The administrators (level
3) are able to access to the scenario editor 242 and the user
manager 246, but cannot view reports from the reporter system 248
or assign training scenarios. The developers (level 4) are able to
access all systems, as well as create new scenarios and coaching
applications.
[0099] The reporter system 248 provides access to a trainee's
simulation results generated by the simulator module 230 and stored
in the database 212, and executes statistical analyses on that
data. The reporter system 248 enables a trainee to review his or
her performance on a particular simulation, review his or her
performance on all simulations over time and view aggregate
performances of trainee groups to compare and contrast the
trainee's progress. For example, in an embodiment of the invention,
the trainee is able to view percentile rankings for the trainee's
statistics for a particular simulation, view aggregate statistics
for all users on selected scenario, view aggregate statistics for
the trainee's team or group, view the trainee's statistics on all
simulations over time, view the team's statistics on one scenario
over time, view the team's statistics on all simulations over time
and view and edit notes attached to particular simulation
results.
[0100] The reporter system 248 includes various result records,
such as the scenario result records, the dialog result records and
the iteration result records. The scenario result records include
the result of a single trainee taking a single scenario on a
particular occasion. Multiple scenario result records exist when a
trainee takes a particular scenario multiple times. The scenario
result record contains data about the trainee's performance
relative to the entire scenario, including the date the scenario
was taken, notes amended by the trainer, the trainee's overall
statistics, and the amount of time spent in the scenario. The
dialog result record is a component of a scenario result record and
stores information about a trainee's performance on a particular
dialog segment of a scenario. The iteration result record is a
component of a dialog result record and includes performance
information about a particular iteration of a dialog segment of a
scenario for a trainee.
[0101] The reporter system 248 is flexibly integrated with the
various dynamic aspects of the training program to assure proper
context and accuracy of the reports. For example, the level of
coaching can be adjusted by the trainer on an individual trainee
basis, as discussed with respect to FIG. 7 above. In other words,
two trainees participating in the same scenario may have different
levels of coaching. It is therefore possible that a more advanced
trainee, who has little to no coaching, performs the same as an
inferior trainee, who has the highest level of coaching. The
dynamic nature of the reporter system 248 factors in that the score
of the more advanced trainee was obtained without the full benefit
of coaching and reports accordingly. The reporter system 248
therefore provides an accurate picture of the training results
customized to the trainee's individual training experience.
[0102] Although the invention has been described with reference to
several exemplary embodiments, it is understood that the words that
have been used are words of description and illustration, rather
than words of limitation. Changes may be made within the purview of
the appended claims, as presently stated and as amended, without
departing from the scope and spirit of the invention in its
aspects. Although the invention has been described with reference
to particular means, materials and embodiments, the invention is
not intended to be limited to the particulars disclosed; rather,
the invention extends to all functionally equivalent structures,
methods and uses such as are within the scope of the appended
claims.
[0103] In accordance with various embodiments of the present
invention, the methods described herein are intended for operation
as software programs running on a computer processor. Dedicated
hardware implementations including, but not limited to, application
specific integrated circuits, programmable logic arrays and other
hardware devices can likewise be constructed to implement the
methods described herein. Furthermore, alternative software
implementations including, but not limited to, distributed
processing or component/object distributed processing, parallel
processing, or virtual machine processing can also be constructed
to implement the methods described herein.
[0104] It should also be noted that the software implementations of
the present invention as described herein are optionally stored on
a tangible storage medium, such as: a magnetic medium such as a
disk or tape; a magneto-optical or optical medium such as a disk;
or a solid state medium such as a memory card or other package that
houses one or more read-only (non-volatile) memories, random access
memories, or other re-writable (volatile) memories. A digital file
attachment to e-mail or other self-contained information archive or
set of archives is considered a distribution medium equivalent to a
tangible storage medium. Accordingly, the invention is considered
to include a tangible storage medium or distribution medium, as
listed herein and including art-recognized equivalents and
successor media, in which the software implementations herein are
stored.
[0105] Although the present specification describes components and
functions implemented in the embodiments with reference to
particular standards and protocols, the invention is not limited to
such standards and protocols. Each of the standards for Internet
and other packet switched network transmission (e.g., TCP/IP,
UDP/IP, HTML, XML, HTTP) represent examples of the state of the
art. Such standards are periodically superseded by faster or more
efficient equivalents having essentially the same functions.
Accordingly, replacement standards and protocols having the same
functions are considered equivalents.
* * * * *